« Public Not With the Bush Program | Main | Andy Stern May Be a Little Bit Crazy, But Sometimes Crazy Is What You Need »

Yes, Virginia, Question Wording Does Matter

The Pew Research Center has just released a useful little study, "Social Security Polling: Cross-Currents in Opinion about Private Accounts". The study looks at a number of different types of questions that have been asked recently about private accounts and the President's Social Security proposal and compares how results differ on a given type of question, depending on question wording. Here are some of the findings, which should probably come as no big surprise to those who have been reading my posts and/or closely following recent polling on the issue.

1. Polls show that the public thinks Social Security faces major problems, but does not believe it is in a "crisis". Moreover, the public is less likely to believe Social Security system needs a complete overhaul or major change than that the health or education systems need such change.

2. Questions which do not mention the risks or tradeoffs involved in setting up private accounts generally get a more positive response than those that do.

3. Among questions that simply mention the general idea of private accounts, those that associate the idea with Bush and/or do not specify that only some of workers' contributions could go into private accounts get the most negative response. (Note: recent DCorps and Hart Research/AFL-CIO results indicate that support for the general idea, both with and without specific reference to Bush, may have declined in the short time since the questions quoted in the Pew study were asked.)

4. Among questions that do refer to the tradeoffs of private accounts, questions that allude directly to risk and, particularly, the cut in guaranteed benefits that accompany these accounts get the most negatives response.

5. Finally, questions that query initial supporters of the general idea whether they would support private accounts if it meant a reduction in guaranteed benefits yield the most negative results of all: a hardcore of support for private accounts in the 22-23 percent range.

There you have it. These varying results suggest that private accounts will not have an easy time of it unless the public forgets that Bush is pushing the accounts and never connects these accounts to risk and a reduction in guaranteed benefits. Such a know-nothing public no longer seems likely on this hotly-debated issue.